Part 1 - refactor utf8 utilities to shared

include/aws/mqtt/mqtt.h

@@ -101,9 +101,18 @@ AWS_EXTERN_C_BEGIN
 
 AWS_MQTT_API
 bool aws_mqtt_is_valid_topic(const struct aws_byte_cursor *topic);
+
 AWS_MQTT_API
 bool aws_mqtt_is_valid_topic_filter(const struct aws_byte_cursor *topic_filter);
 
+/**
+ * Validate utf-8 string under mqtt specs
+ *
+ * @param text
+ * @return AWS_OP_SUCCESS if the text is validate, otherwise AWS_OP_ERR
+ */
+AWS_MQTT_API int aws_mqtt_validate_utf8_text(struct aws_byte_cursor text);
+
 /**
  * Initializes internal datastructures used by aws-c-mqtt.
  * Must be called before using any functionality in aws-c-mqtt.

include/aws/mqtt/private/v5/mqtt5_utils.h

@@ -94,14 +94,6 @@ AWS_EXTERN_C_BEGIN
  */
 AWS_MQTT_API extern struct aws_byte_cursor g_aws_mqtt5_connect_protocol_cursor;
 
-/**
- * Validate utf-8 string under mqtt5 specs
- *
- * @param text
- * @return AWS_OP_SUCCESS if the text is validate, otherwise AWS_OP_ERR
- */
-AWS_MQTT_API int aws_mqtt5_validate_utf8_text(struct aws_byte_cursor text);
-
 /**
  * Simple helper function to compute the first byte of an MQTT packet encoding as a function of 4 bit flags
  * and the packet type.

include/aws/mqtt/v5/mqtt5_client.h

@@ -787,18 +787,6 @@ AWS_MQTT_API int aws_mqtt5_negotiated_settings_init(
     struct aws_mqtt5_negotiated_settings *negotiated_settings,
     const struct aws_byte_cursor *client_id);
 
-/**
- * Makes an owning copy of a negotiated settings structure
- *
- * @param source settings to copy from
- * @param dest settings to copy into.  Must be in a zeroed or initialized state because it gets clean up
- *  called on it as the first step of the copy process.
- * @return success/failure
- */
-AWS_MQTT_API int aws_mqtt5_negotiated_settings_copy(
-    const struct aws_mqtt5_negotiated_settings *source,
-    struct aws_mqtt5_negotiated_settings *dest);
-
 /**
  * Clean up owned memory in negotiated_settings
  *

source/mqtt.c

@@ -5,8 +5,8 @@
 
 #include <aws/mqtt/mqtt.h>
 
+#include <aws/common/encoding.h>
 #include <aws/http/http.h>
-
 #include <aws/io/logging.h>
 
 /*******************************************************************************
@@ -289,3 +289,37 @@ void aws_mqtt_fatal_assert_library_initialized(void) {
         AWS_FATAL_ASSERT(s_mqtt_library_initialized);
     }
 }
+
+/* UTF-8 encoded string validation respect to [MQTT-1.5.3-2]. */
+static int aws_mqtt_utf8_decoder(uint32_t codepoint, void *user_data) {
+    (void)user_data;
+    /* U+0000 - A UTF-8 Encoded String MUST NOT include an encoding of the null character U+0000. [MQTT-1.5.4-2]
+     * U+0001..U+001F control characters are not valid
+     */
+    if (AWS_UNLIKELY(codepoint <= 0x001F)) {
+        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
+    }
+
+    /* U+007F..U+009F control characters are not valid */
+    if (AWS_UNLIKELY((codepoint >= 0x007F) && (codepoint <= 0x009F))) {
+        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
+    }
+
+    /* Unicode non-characters are not valid: https://www.unicode.org/faq/private_use.html#nonchar1 */
+    if (AWS_UNLIKELY((codepoint & 0x00FFFF) >= 0x00FFFE)) {
+        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
+    }
+    if (AWS_UNLIKELY(codepoint >= 0xFDD0 && codepoint <= 0xFDEF)) {
+        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
+    }
+
+    return AWS_ERROR_SUCCESS;
+}
+
+static struct aws_utf8_decoder_options s_aws_mqtt_utf8_decoder_options = {
+    .on_codepoint = aws_mqtt_utf8_decoder,
+};
+
+int aws_mqtt_validate_utf8_text(struct aws_byte_cursor text) {
+    return aws_decode_utf8(text, &s_aws_mqtt_utf8_decoder_options);
+}

source/v5/mqtt5_options_storage.c

@@ -172,7 +172,7 @@ static int s_aws_mqtt5_user_property_set_validate(
             return aws_raise_error(AWS_ERROR_MQTT5_USER_PROPERTY_VALIDATION);
         }
 
-        if (aws_mqtt5_validate_utf8_text(property->name)) {
+        if (aws_mqtt_validate_utf8_text(property->name)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL, "id=%p: %s - user property #%zu name not valid UTF8", log_context, log_prefix, i);
             return aws_raise_error(AWS_ERROR_MQTT5_USER_PROPERTY_VALIDATION);
@@ -187,7 +187,7 @@ static int s_aws_mqtt5_user_property_set_validate(
                 property->value.len);
             return aws_raise_error(AWS_ERROR_MQTT5_USER_PROPERTY_VALIDATION);
         }
-        if (aws_mqtt5_validate_utf8_text(property->value)) {
+        if (aws_mqtt_validate_utf8_text(property->value)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: %s - user property #%zu value not valid UTF8",
@@ -332,7 +332,7 @@ int aws_mqtt5_packet_connect_view_validate(const struct aws_mqtt5_packet_connect
         return aws_raise_error(AWS_ERROR_MQTT5_CONNECT_OPTIONS_VALIDATION);
     }
 
-    if (aws_mqtt5_validate_utf8_text(connect_options->client_id)) {
+    if (aws_mqtt_validate_utf8_text(connect_options->client_id)) {
         AWS_LOGF_ERROR(
             AWS_LS_MQTT5_GENERAL,
             "id=%p: aws_mqtt5_packet_connect_view - client id not valid UTF-8",
@@ -349,7 +349,7 @@ int aws_mqtt5_packet_connect_view_validate(const struct aws_mqtt5_packet_connect
             return aws_raise_error(AWS_ERROR_MQTT5_CONNECT_OPTIONS_VALIDATION);
         }
 
-        if (aws_mqtt5_validate_utf8_text(*connect_options->username)) {
+        if (aws_mqtt_validate_utf8_text(*connect_options->username)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: aws_mqtt5_packet_connect_view - username not valid UTF-8",
@@ -1259,7 +1259,7 @@ int aws_mqtt5_packet_disconnect_view_validate(const struct aws_mqtt5_packet_disc
             return aws_raise_error(AWS_ERROR_MQTT5_DISCONNECT_OPTIONS_VALIDATION);
         }
 
-        if (aws_mqtt5_validate_utf8_text(*disconnect_view->reason_string)) {
+        if (aws_mqtt_validate_utf8_text(*disconnect_view->reason_string)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: aws_mqtt5_packet_disconnect_view - reason string not valid UTF-8",
@@ -1591,7 +1591,7 @@ int aws_mqtt5_packet_publish_view_validate(const struct aws_mqtt5_packet_publish
         AWS_LOGF_ERROR(
             AWS_LS_MQTT5_GENERAL, "id=%p: aws_mqtt5_packet_publish_view - missing topic", (void *)publish_view);
         return aws_raise_error(AWS_ERROR_MQTT5_PUBLISH_OPTIONS_VALIDATION);
-    } else if (aws_mqtt5_validate_utf8_text(publish_view->topic)) {
+    } else if (aws_mqtt_validate_utf8_text(publish_view->topic)) {
         AWS_LOGF_ERROR(
             AWS_LS_MQTT5_GENERAL, "id=%p: aws_mqtt5_packet_publish_view - topic not valid UTF-8", (void *)publish_view);
         return aws_raise_error(AWS_ERROR_MQTT5_PUBLISH_OPTIONS_VALIDATION);
@@ -1626,7 +1626,7 @@ int aws_mqtt5_packet_publish_view_validate(const struct aws_mqtt5_packet_publish
 
         // Make sure the payload data is UTF-8 if the payload_format set to UTF8
         if (*publish_view->payload_format == AWS_MQTT5_PFI_UTF8) {
-            if (aws_mqtt5_validate_utf8_text(publish_view->payload)) {
+            if (aws_mqtt_validate_utf8_text(publish_view->payload)) {
                 AWS_LOGF_ERROR(
                     AWS_LS_MQTT5_GENERAL,
                     "id=%p: aws_mqtt5_packet_publish_view - payload value is not valid UTF-8 while payload format "
@@ -1646,7 +1646,7 @@ int aws_mqtt5_packet_publish_view_validate(const struct aws_mqtt5_packet_publish
             return aws_raise_error(AWS_ERROR_MQTT5_PUBLISH_OPTIONS_VALIDATION);
         }
 
-        if (aws_mqtt5_validate_utf8_text(*publish_view->response_topic)) {
+        if (aws_mqtt_validate_utf8_text(*publish_view->response_topic)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: aws_mqtt5_packet_publish_view - response topic not valid UTF-8",
@@ -1692,7 +1692,7 @@ int aws_mqtt5_packet_publish_view_validate(const struct aws_mqtt5_packet_publish
             return aws_raise_error(AWS_ERROR_MQTT5_PUBLISH_OPTIONS_VALIDATION);
         }
 
-        if (aws_mqtt5_validate_utf8_text(*publish_view->content_type)) {
+        if (aws_mqtt_validate_utf8_text(*publish_view->content_type)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: aws_mqtt5_packet_publish_view - content type not valid UTF-8",
@@ -2332,7 +2332,7 @@ int aws_mqtt5_packet_unsubscribe_view_validate(const struct aws_mqtt5_packet_uns
 
     for (size_t i = 0; i < unsubscribe_view->topic_filter_count; ++i) {
         const struct aws_byte_cursor *topic_filter = &unsubscribe_view->topic_filters[i];
-        if (aws_mqtt5_validate_utf8_text(*topic_filter)) {
+        if (aws_mqtt_validate_utf8_text(*topic_filter)) {
             AWS_LOGF_ERROR(
                 AWS_LS_MQTT5_GENERAL,
                 "id=%p: aws_mqtt5_packet_unsubscribe_view - topic filter not valid UTF-8: \"" PRInSTR "\"",
@@ -2603,7 +2603,7 @@ static int s_aws_mqtt5_validate_subscription(
     const struct aws_mqtt5_subscription_view *subscription,
     void *log_context) {
 
-    if (aws_mqtt5_validate_utf8_text(subscription->topic_filter)) {
+    if (aws_mqtt_validate_utf8_text(subscription->topic_filter)) {
         AWS_LOGF_ERROR(
             AWS_LS_MQTT5_GENERAL,
             "id=%p: aws_mqtt5_packet_subscribe_view - topic filter \"" PRInSTR "\" not valid UTF-8 in subscription",

source/v5/mqtt5_utils.c

@@ -7,7 +7,6 @@
 
 #include <aws/common/byte_buf.h>
 #include <aws/common/device_random.h>
-#include <aws/common/encoding.h>
 #include <inttypes.h>
 
 uint8_t aws_mqtt5_compute_fixed_header_byte1(enum aws_mqtt5_packet_type packet_type, uint8_t flags) {
@@ -124,24 +123,6 @@ int aws_mqtt5_negotiated_settings_init(
     return AWS_OP_SUCCESS;
 }
 
-int aws_mqtt5_negotiated_settings_copy(
-    const struct aws_mqtt5_negotiated_settings *source,
-    struct aws_mqtt5_negotiated_settings *dest) {
-    aws_mqtt5_negotiated_settings_clean_up(dest);
-
-    *dest = *source;
-    AWS_ZERO_STRUCT(dest->client_id_storage);
-
-    if (source->client_id_storage.allocator != NULL) {
-        return aws_byte_buf_init_copy_from_cursor(
-            &dest->client_id_storage,
-            source->client_id_storage.allocator,
-            aws_byte_cursor_from_buf(&source->client_id_storage));
-    }
-
-    return AWS_OP_SUCCESS;
-}
-
 int aws_mqtt5_negotiated_settings_apply_client_id(
     struct aws_mqtt5_negotiated_settings *negotiated_settings,
     const struct aws_byte_cursor *client_id) {
@@ -553,37 +534,3 @@ bool aws_mqtt_is_topic_filter_shared_subscription(struct aws_byte_cursor topic_c
 
     return true;
 }
-
-/* UTF-8 encoded string validation respect to [MQTT-1.5.3-2]. */
-static int aws_mqtt5_utf8_decoder(uint32_t codepoint, void *user_data) {
-    (void)user_data;
-    /* U+0000 - A UTF-8 Encoded String MUST NOT include an encoding of the null character U+0000. [MQTT-1.5.4-2]
-     * U+0001..U+001F control characters are not valid
-     */
-    if (AWS_UNLIKELY(codepoint <= 0x001F)) {
-        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
-    }
-
-    /* U+007F..U+009F control characters are not valid */
-    if (AWS_UNLIKELY((codepoint >= 0x007F) && (codepoint <= 0x009F))) {
-        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
-    }
-
-    /* Unicode non-characters are not valid: https://www.unicode.org/faq/private_use.html#nonchar1 */
-    if (AWS_UNLIKELY((codepoint & 0x00FFFF) >= 0x00FFFE)) {
-        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
-    }
-    if (AWS_UNLIKELY(codepoint >= 0xFDD0 && codepoint <= 0xFDEF)) {
-        return aws_raise_error(AWS_ERROR_MQTT5_INVALID_UTF8_STRING);
-    }
-
-    return AWS_ERROR_SUCCESS;
-}
-
-struct aws_utf8_decoder_options g_aws_mqtt5_utf8_decoder_options = {
-    .on_codepoint = aws_mqtt5_utf8_decoder,
-};
-
-int aws_mqtt5_validate_utf8_text(struct aws_byte_cursor text) {
-    return aws_decode_utf8(text, &g_aws_mqtt5_utf8_decoder_options);
-}

tests/CMakeLists.txt

@@ -4,7 +4,7 @@ include(AwsLibFuzzer)
 enable_testing()
 
 file(GLOB TEST_HDRS "v3/*.h v5/*.h")
-set(TEST_SRC v3/*.c v5/*.c)
+set(TEST_SRC v3/*.c v5/*.c *.c)
 file(GLOB TESTS ${TEST_HDRS} ${TEST_SRC})
 
 add_test_case(mqtt_packet_puback)
@@ -112,7 +112,7 @@ add_test_case(mqtt5_topic_get_segment_count)
 add_test_case(mqtt5_shared_subscription_validation)
 
 # utf8 utility
-add_test_case(mqtt5_utf8_encoded_string_test)
+add_test_case(mqtt_utf8_encoded_string_test)
 
 # topic aliasing
 add_test_case(mqtt5_inbound_topic_alias_register_failure)